© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 28286
    Prev
    Next
    INDEX
    Explanations

    paying for content

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron responds to words denoting monetary or other exchanges—terms used in contexts of paying, activating (a paid feature), influence (as in quid pro quo), bribes, etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Logical
    -0.85
    莖
    -0.84
     zust
    -0.82
     Dirty
    -0.80
    借り
    -0.79
    adio
    -0.77
     Inv
    -0.77
    校验
    -0.76
    的原因
    -0.75
     prototype
    -0.75
    POSITIVE LOGITS
     overpriced
    1.01
    premium
    0.93
     pembeli
    0.92
    arza
    0.92
     bénéfices
    0.90
     services
    0.90
     premium
    0.89
     tjen
    0.89
     émotions
    0.88
    gewöhn
    0.87
    Activations Density 0.073%

    No Known Activations